He guys,
I have an idea for a software, and im pretty sure that somebody did something like this already.
In short: Clustering together loads of pcs to create a deep kibitzer.
A bit longer: You kibitz a match (the idea came to me when watching tcec, together with a 100 other ppl, all wondering what was going on). You run an engine, select candidate moves and split: Queue every candidate move. The clients can work together on the queue, everyone using the same engine, analyze the position till say depth 30, and post the moves. You would build a tree pretty fast, and cut it as the game progresses. So essentially a shallow search with a very deep analysis.
Thinking about it you could actually just use the search algo of an existing engine, just replace the eval...
Something like this HAS TO exist, does it?
Cluster Chess Analysis
-
User923005
- Posts: 616
- Joined: Thu May 19, 2011 1:35 am
Re: Cluster Chess Analysis
Yes it does, but it is not very efficient.
The network is not nearly so fast as the machine's bus.
Anyway, an ancient thing that did what you ask is called "chessbrain".
http://chessbrain.net/
If you examine Daniel Shawul's program Scorpio, you will find that it is cluster capable, right out of the box.
https://github.com/dshawul/Scorpio
The network is not nearly so fast as the machine's bus.
Anyway, an ancient thing that did what you ask is called "chessbrain".
http://chessbrain.net/
If you examine Daniel Shawul's program Scorpio, you will find that it is cluster capable, right out of the box.
https://github.com/dshawul/Scorpio